A link to improve stroke patient care: a successful linkage between a statewide emergency medical services data system and a stroke registry.

OBJECTIVES: regionalization of stroke care, including diversion to stroke centers, requires that emergency medical services (EMS) systems accurately identify acute stroke patients. A barrier to evaluating and improving EMS stroke patient identification is the inability to link EMS data with hospital data for individual patients. We sought to create and validate a linkage of the North Carolina EMS Data System (NC-EMS-DS) with data contained in the North Carolina Stroke Care Collaborative (NCSCC) Registry. METHODS: all NCSCC Registry patients arriving to one of three hospitals by EMS in a 6-month period were matched against NC-EMS-DS. Records were deterministically matched on receiving hospital, hospital arrival date/time, age, and sex. We performed linkage validation by providing each site investigator with a stroke patient list derived from North Carolina Stroke Care Collaborative Registry (NC-EMS-DS), matched by individual patient to deidentified data in the NCSCCR. Each site investigator determined the set of true matches by comparing the matched list to a NCSCCR patient identifier key maintained at each site. Incorrect matches were reviewed by the research team to identify methods for future improvement in the matching logic. RESULTS: for the three validation hospitals, 753 NCSCC Registry patients arrived by EMS. For these patients, 473 (63%) matches to local EMS records were identified, and 421 (89%) of the matches were verified using full patient identifiers. Most match verification failures were due to incorrect date/time stamp and inability to find a corresponding EMS record. CONCLUSIONS: linking EMS records electronically to a stroke registry is feasible and leads to a large number of valid matches. This small validation is limited by EMS data quality. Matching may improve with better EMS documentation and standardized facility documentation.

Duke Scholars

Author Emily O'Brien Population Health Sciences